B ROAD-BAND MICROWAVE NOISE SOURCE 



613 



match at any operating frequency from MOi) mc U) 4500 mc. The admittance 

 diagram when the circuit was adjusted for match at .^960 mc is shown in 

 Fig. 3; the standing wave ratio was less than 2.9 db from v^70() to 4240 mc. 



At 3960 mc the conductance of the gaseous discharge varied cUrectly with 

 the (Urect current, while the negative susceptance had a broad maximum of 

 —j.bl I'o mhos at a current of 65 to 100 milhamperes, as shown in Fig. 4. 

 These values are for the gaseous discharge; the susceptances of the enclosing 

 glass tubing, the back piston and the holes in the sidewalls have been sub- 

 tracted from the measured results. It is interesting to note that the discharge 

 appears to be inductive. 



The waveguide circuit containing the gaseous discharge tube was con- 

 nected to the input waveguide of a sensitive microwave receiver which was 

 used as a relative noise power meter. The noise power available from the 



Fig. 2 — Waveguide circuit for microwave noise generator using a gaseous discharge 



tube. 



gaseous discharge was substantially independent of the direct current from 



40 ma to 140 ma. These data are plotted in Fig. 5, which gives 10 log ( ^ — 1 j 



versus direct current in milliamperes. The ordinate has been chosen so as 

 to conform witli absolute measurements made subsequently. The r.m.s. 

 deviation from the straight line which represents a probable coefficient of 

 only —.003 db per milliampere was about ±.05 db. We do not claim to be 

 able to achieve even this degree of accuracy with our present measuring 

 equipment and hence do not place much coniidence in the numerical value 

 of this coefficient. Actually the decrease in noise with increasing current 

 may have been associated with a change in the ambient temperature rather 

 than with the increased current density. At least it is in the right direction 

 for this to be the case. 



The temperature coefficient of the noise from the discharge was found to 

 be negative; when a piece of dry ice was held on the tubular shield of the 

 circuit for a few minutes (long enough for frost to form on the brass) the 

 output noise power of the discharge increased 0.6 db. The circuit was heated 



